Default

Monthly account scoring that tells reps who to call

Every account is scored on fit and timing each month, so reps know which accounts to call now and which get automated emails

Nandika Jhunjhunwala Founding Growth Engineer, Default LinkedIn

The problem

Buying signals like hiring, funding and M&A were easy to get but sat in isolation. With no way to combine them at scale, reps fell back on manual research or gut feel, there was no single record that BDRs, AEs, marketing and the email-sending setup all worked from, and high-fit, in-market accounts were being missed.
What they built

An account scoring system that pulls every signal Default cares about for a company, stores it in one place, and scores the account on two questions: is this the right kind of buyer (fit), and is it changing right now in ways that suggest it's ready to buy (timing). Accounts land in four tiers: P1 to call now, P2 for email and nurture, P3 for automated low-touch sequences, and P4 to ignore. A web app shows each account's data and lets the team score one-off accounts. Nandika built it with her manager acting as product manager; she says what was scoped as a 6 to 12 month project for a team of engineers took about a month with no engineers.
So what

Every rep knows which accounts to call this month and which get automated email. Because the system records the score at which meetings were booked, the scoring can be checked against results.

Under the hood

Accounts come in four ways: bulk CSV import, one-off scoring in the web app, Salesforce sync, and accounts that engage through website visits, form fills, email, LinkedIn or ads. They are enriched with Clearbit and other providers, stored in a 12-table Supabase Postgres database, and orchestrated with GitHub Actions. Fit covers firmographics and technographics (headcount, vertical, funding, growth stage, RevOps and GTM engineering presence, CRM setup); timing covers change signals. Job descriptions are stored and run through an LLM to infer signals such as AI initiatives or pains Default solves. Hard disqualifiers (e.g. B2C) apply. Scores write to a custom Salesforce field. Nandika chose a database over Clay because Clay suits one-time enrichment, not scoring a whole TAM, and a database gives audit logs and snapshots. Outcome data (score at meeting, touchpoints to convert) is stored to refine the model later.